TitleIFN-β-inducing, unusual viral RNA species produced byparamyxovirus infection accumulated into distinct cytoplasmicstructures in an RNA-type-dependent manner

Author(s) Yoshida, Asuka; Kawabata, Ryoko; Honda, Tomoyuki;Tomonaga, Keizo; Sakaguchi, Takemasa; Irie, Takashi

Citation Frontiers in Microbiology (2015), 6

Issue Date 2015-08-04

URL http://hdl.handle.net/2433/215934

Right

© 2015 Yoshida, Kawabata, Honda, Tomonaga, Sakaguchi andIrie. This is an open-access article distributed under the termsof the Creative Commons Attribution License (CC BY). Theuse, distribution or reproduction in other forums is permitted,provided the original author(s) or licensor are credited and thatthe original publication in this journal is cited, in accordancewith accepted academic practice. No use, distribution orreproduction is permitted which does not comply with theseterms.

Type Journal Article

Textversion publisher

Kyoto University

ORIGINAL RESEARCHpublished: 04 August 2015

doi: 10.3389/fmicb.2015.00804

Edited by:Nobuhiro Suzuki,

Okayama University, Japan

Reviewed by:Takasuke Fukuhara,

Osaka University, JapanHiromichi Dansako,

Okayama University, Japan

*Correspondence:Takashi Irie and

Takemasa Sakaguchi,Department of Virology, Institute

of Biomedical and Health Sciences,Hiroshima University, 1-2-3 Kasumi,

Minami-ku, Hiroshima 734-8551,Japan

tirie@hiroshima-u.ac.jp;tsaka@hiroshima-u.ac.jp

Specialty section:This article was submitted to

Virology,a section of the journal

Frontiers in Microbiology

Received: 08 June 2015Accepted: 22 July 2015

Published: 04 August 2015

Citation:Yoshida A, Kawabata R, Honda T,

Tomonaga K, Sakaguchi T and Irie T(2015) IFN-β-inducing, unusual viral

RNA species produced byparamyxovirus infection accumulated

into distinct cytoplasmic structuresin an RNA-type-dependent manner.

Front. Microbiol. 6:804.doi: 10.3389/fmicb.2015.00804

IFN-β-inducing, unusual viral RNAspecies produced by paramyxovirusinfection accumulated into distinctcytoplasmic structures in anRNA-type-dependent mannerAsuka Yoshida1, Ryoko Kawabata1, Tomoyuki Honda2, Keizo Tomonaga2,Takemasa Sakaguchi1* and Takashi Irie1*

1 Department of Virology, Institute of Biomedical and Health Sciences, Hiroshima University, Hiroshima, Japan, 2 Departmentof Viral Oncology, Institute for Virus Research, Kyoto University, Kyoto, Japan

The interferon (IFN) system is one of the most important defensive responses ofmammals against viruses, and is rapidly evoked when the pathogen-associatedmolecular patterns (PAMPs) of viruses are sensed. Non-self, virus-derived RNA specieshave been identified as the PAMPs of RNA viruses. In the present study, we compareddifferent types of IFN-β-inducing and -non-inducing viruses in the context of Sendaivirus infection. We found that some types of unusual viral RNA species were producedby infections with IFN-β-inducing viruses and accumulated into distinct cytoplasmicstructures in an RNA-type-dependent manner. One of these structures was similar tothe so-called antiviral stress granules (avSGs) formed by an infection with IFN-inducingviruses whose C proteins were knocked-out or mutated. Non-encapsidated, unusualviral RNA harboring the 5′-terminal region of the viral genome as well as RIG-I andtypical SG markers accumulated in these granules. Another was a non-SG-like inclusionformed by an infection with the Cantell strain; a copyback-type DI genome, but not anauthentic viral genome, specifically accumulated in the inclusion, whereas RIG-I and SGmarkers did not. The induction of IFN-β was closely associated with the production ofthese unusual RNAs as well as the formation of the cytoplasmic structures.

Keywords: innate immunity and responses, pathogen-associated molecular patterns (PAMPs), stress granules,Sendai virus (SeV), RNA Viruses

Introduction

Eukaryotic cells are equipped with various defense mechanisms to detect and respond to viralinfections rapidly. The interferon (IFN) system is one of the most important natural defensesof mammalian cells in the early phase of viral infection. Host cells sense the invasion of virusesby recognizing their pathogen-associated molecular patterns (PAMPs), including the structuralcharacteristics of viral RNAs that differentiate them from cellular RNAs (Akira et al., 2006). ViralRNAs are detected by non-self RNA sensors such as Toll-like receptors (TLR), and a family ofcytosolic RNA helicases termed RIG-I-like receptors (RLRs), including retinoic-acid induciblegene-I (RIG-I), melanoma differentiation-associated gene 5 (MDA5), and laboratory of physiology

Frontiers in Microbiology | www.frontiersin.org 1 August 2015 | Volume 6 | Article 804

Yoshida et al. Selective accumulation of IFN-β-inducing viral RNA

and genetics gene 2 (LGP2). This is followed by the subsequentinduction of IFN-β (Gitlin et al., 2006; Kaisho and Akira, 2006;Saito et al., 2007). Autocrine or paracrine IFNs bind to IFNreceptors on the cell surface, leading to the expression of 100sof IFN-stimulated genes (ISGs) through the Jak/STAT signalingpathway, which ultimately exerts various antiviral effects (Akiraet al., 2006).

Translational arrest is one of the IFN responses of hostcells triggered by viral infections. Various eukaryotic translationinitiation factor 2 (eIF2) kinases such as protein kinase R (PKR)are activated in response to IFNs, and the accumulation ofphosphorylated eIF2α inhibits the translation of both cellularand viral mRNAs (Anderson and Kedersha, 2002; Kedershaand Anderson, 2002; Holcik and Sonenberg, 2005). Cytoplasmicstress granules (SGs), which are the foci of concentrated48S translation preinitiation complexes and defined by certainmarker RNA binding proteins such as T-cell intracellularantigen-1 (TIA-1), TIA-1-related protein (TIAR), and Ras-Gap-SH3 domain-binding protein (G3BP1), are formed underthese conditions (Anderson and Kedersha, 2002; Kedersha andAnderson, 2002). Because they contain stable inert mRNA,SGs are believed to serve as temporary sites at which mRNA-protein complexes are stored to pause active translation orbe decayed in adjacent processing bodies (Anderson andKedersha, 2006; Balagopal and Parker, 2009; Buchan and Parker,2009).

A number of viruses have been shown to induce theformation of SGs in infected cells, and this may be relatedto the virus-induced shut-off of cellular protein translation.For example, hepatitis C virus, poliovirus, Semliki Forest virus,and mammalian orthoreovirus promote the shut-off of cellularproteins and the assembly of SGs at the early phase of infection,and this is inhibited as the infection progresses (McInerney et al.,2005; White et al., 2007; Qin et al., 2009, 2011; Ariumi et al.,2011; White and Lloyd, 2011; Garaigorta et al., 2012; Panas et al.,2012; Ruggieri et al., 2012; Fitzgerald and Semler, 2013; Pageret al., 2013; Carroll et al., 2014). Two RNA viruses, respiratorysyncytial virus and coronavirus, are also known to utilize SGs aspart of the machinery to inhibit host cellular protein translation(Raaben et al., 2007; Lindquist et al., 2010, 2011). SGs were notdetected during infection by influenza A virus (IAV); however,a recombinant IAV lacking non-structural protein 1 (NS1), aninhibitor of PKR, efficiently induced the formation of SGs and theproduction of IFN-β in a PKR-dependent manner (Khaperskyyet al., 2012; Mok et al., 2012; Onomoto et al., 2012). In this case,SGs were suggested to play an important role as the sites of viralRNA sensing and subsequent anti-viral responses, because RLRslocalized together with viral nucleoproteins and RNA as well asanti-viral proteins in SGs (Onomoto et al., 2012; Ng et al., 2013).

The RNA species produced during the course of RNA viralreplication, such as mRNA, dsRNA, and 5′-triphosphate (5′-ppp)RNA including leader, trailer, genome, and antigenome RNAs aswell as defective interfering (DI) genomes, have been shown totrigger the production of type I IFNs (Yoneyama et al., 2004;Hornung et al., 2006; Pichlmair et al., 2006; Strahle et al., 2007;Hausmann et al., 2008; Baum et al., 2010; Baum and Garcia-Sastre, 2011; Kato et al., 2011; Marq et al., 2011; Davis et al.,

2012; Bowzard et al., 2013; Weber et al., 2013; Runge et al.,2014; Schmolke et al., 2014). We and other groups have recentlyreported that recombinant viruses of Sendai virus (SeV), aprototype of the family Paramyxoviridae, in which the C proteinsare knocked-out or mutated, generate dsRNA in infected cells atlevels similar to the production of IFN-β (Takeuchi et al., 2008;Irie et al., 2010). Previous studies also reported that, in the casesof SeV and IAV, copyback (cb)- and internal deletion (id)-typeDI genomes, respectively, rather than full-length viral genomes,preferentially associated with RIG-I and strongly induced theproduction of IFN-β (Baum et al., 2010; Baum and Garcia-Sastre,2011).

In spite of the large number of studies conducted in thisfield, it remains unknown what kinds of viral RNA speciesare recognized by RLRs and where the sites of recognition arein real infections by RNA viruses. In the present study, wecompared some types of IFN-β-non-inducing and IFN-β-highlyinducing viruses in the context of SeV infection. One of thebiggest advantages of this study is that the comparison canbe performed within the context of the same viral species.We found that some types of unusual RNA species that weredistinguishable according to specific detectability by the anti-dsRNA antibody or FISH analysis were produced during the viralreplication of IFN-inducing SeVs, but not IFN-non-inducing SeVstrains, and accumulated into distinct cytoplasmic structures inan RNA-type-dependent manner. These unusual RNAs exhibiteddistinct properties in infected cells in terms of encapsidationwith the viral N protein and subcellular distribution with SGmarker proteins and RLRs. Our results suggest that RNA-type-dependent mechanisms recognize and accumulate virus-derived,IFN-β-inducible, unusual RNAs into specific compartment totrigger the production of IFN-β, and that SeV may evadedetection by the host innate immune system by preventing theproduction of these RNA species.

Materials and Methods

Cells, Viruses, and PlasmidsLLC-MK2 cells (macaquemonkey kidney-derived cells, describedin Kiyotani et al., 1990) and HeLa cells (CCL-2; purchased fromATCC) were maintained as described previously (Irie et al.,2012). All of the SeVs, even the WT of strain Z and Hamamatsu(HMT), used in this study were recovered from cDNA using areverse genetics technique as described previously (Kato et al.,1996; Fujii et al., 2002), except for the Cantell (CNT; VR-907;ATCC), Fushimi (FSM), and Nagoya (NGY) strains. The SeVrecombinants, C′/C(-), 4C(-), and V(-), were kindly provided byKato (National Institute of Infectious Diseases, Japan; Kato et al.,1997; Kurotani et al., 1998). All of the SeVs as well as virulentand avirulent Newcastle disease virus (NDV) Miyadera and D26strains (Toyoda et al., 1987), respectively, were propagated inembryonated chicken eggs. SeV and NDV titers were determinedby an immunofluorescent infectious focus assay in LLC-MK2cells and expressed as cell infectious units (CIUs)/ml, as describedpreviously (Kiyotani et al., 1990). One-step growth kinetics of theviruses was determined as described previously (Irie et al., 2014).

Frontiers in Microbiology | www.frontiersin.org 2 August 2015 | Volume 6 | Article 804

Yoshida et al. Selective accumulation of IFN-β-inducing viral RNA

Plasmids encoding the P, C, and V proteins of SeV strain Z in thepCAGGS.MCS vector have been described previously (Sakaguchiet al., 2005, 2011; Irie et al., 2008b, 2012).

AntibodiesThe polyclonal antibodies (pAbs) against whole virions of SeVand NDV were described previously (Kiyotani et al., 1990). ThepAbs against the SeV P and C proteins were kindly provided byKato. The monoclonal antibody (mAb) against the SeVN proteinwas kindly provided by Suzuki (National Institute of InfectiousDisease, Japan). The mAb and pAb against G3BP1 (sc-365338,Santa Cruz Biotechnology; and ab39533, Abcam, respectively),pAb against RIG-I (28137; Immuno-Biological Laboratories,Japan), mAb against TIAR, and dsRNA (#8509, Cell SignalingTechnology; and J2, Scicons, Hungary, respectively) were usedaccording to the protocols of the suppliers.

Sodium Arsenite and IFN-α TreatmentHeLa cells cultured on glass coverslips were infected with theindicated viruses or transfected with the indicated plasmids. Theculture medium was replaced by serum-free DMEM 24 h post-infection (p.i.) or post-transfection (p.t.), and cells were treatedwith sodium arsenite (NaAsO2; Sigma) at a final concentration of0.5 mM for 30 min or with IFN-α (1,000 IU/ml; R&D Systems)for 6 h.

Immunofluorescence MicroscopyHeLa cells cultured on glass coverslips were transfected withthe indicated plasmids using FuGENE HD transfection reagent(Promega) or infected with the indicated viruses at an MOI of5. At 24 h p.t. or p.i., cells were fixed and immunostained asdescribed previously using appropriate combinations of primaryand secondary antibodies (Irie et al., 2013). To detect RIG-I, anddsRNA, the Tyramide signal amplification (TSA) kit with HRP-Goat Anti-Mouse IgG and Alexa Fluor 488 Tyramide (MolecularProbes) were used to increase the detection sensitivity. Coverslipswere mounted on glass slides with the SlowFade Gold antifadereagent with or without DAPI (Molecular Probes) and observedusing an LSM 5 confocal microscope (Carl Zeiss).

RNA PreparationHeLa cells were infected with the indicated viruses at an MOI of5. At 24 h p.i., total RNA was prepared using the High Pure RNAIsolation Kit (Roche Diagnostics). Viral RNA in the working viralstocks was prepared using the High Pure Viral RNA Kit (RocheDiagnostics).

Quantitative RT-PCRQuantitative (q) RT-PCR was performed as described previously(Irie et al., 2010, 2014). qRT-PCR samples were analyzedusing the Eco Real-Time PCR System (Illumina). For directcomparison, all of the indicated samples were analyzed in thesame experiments. To detect the genome-length viral RNAsand cbDI genomes of SeV stocks, qRT-PCR was performedusing the primer sets of 5SeVZ1683 + 3SeVZ1843, as describedpreviously (Irie et al., 2008a, 2014), and 5cbDIdetect15,312-15,293 + 3cbDIdetect15,033-15,014, which were complementary

to the regions of the indicated positions of SeV genome RNA, asdescribed recently by Baum et al. (2010).

ImmunoprecipitationHeLa cells were lysed in RIPA buffer (0.5% NP-40, 20 mM Tris-HCl [pH 7.4], 150 mM NaCl) after 24 h of infection by theindicated viruses or 30 min of the arsenite treatment, and theinsoluble fraction was removed by high-speed centrifugation. Theviral proteins in the lysates were removed by three consecutiveimmunoprecipitation steps using anti-SeV or anti-NDV pAbsand Protein G Sepharose beads (GE Healthcare Life Sciences).Supernatants were harvested from the final immunoprecipitationsamples, and were then subjected to RNA preparation, asdescribed above.

RNA TransfectionHeLa cells cultured on glass coverslips were transfected with1 μg of the indicated RNA samples prepared above, togetherwith 0.5 μg of an empty pUC19 vector, using the FuGENE HDtransfection reagent.

Preparation of the CNT-lowDI Viral StockCantell samples (1.4 × 109 CIU/ml) that were 108∼9-fold seriallydiluted were inoculated into the allantoic cavity of 10-days-oldembryonated chicken eggs, and incubated for 72 h at 34◦C.Allantoic fluid was harvested from the eggs, and the titers of eachfluid stock were determined as described above. RNA sampleswere prepared from 100 μl of each fluid stock, and the ratios ofthe cbDI genomes to viral genomes were determined by qRT-PCRas described above.

Fluorescence In Situ Hybridization (FISH)HeLa cells cultured on glass coverslips were infected withthe indicated viruses. At 24 h p.i., cells were fixed with 3%paraformaldehyde solution in PBS, and then subjected to FISHanalysis using the FISH Tag RNA Green Kit with Alexa Fluor 488dye (Molecular Probes) according to the protocol of the supplier.The RNA probe was designed to be complementary to theregion of 14,761–15,384 in SeV genome RNA. After FISH, somesamples were further subjected to fluorescent immunodetectionas described above using the indicated antibodies. Final sampleswere observed using an LSM 5 confocal microscope.

Results

Strong Correlations between Levels of theInduction of IFN-β and the Formation ofG3BP1-Positive Granules by Infection ofC-Mutated and -Deficient SeV RecombinantsWe first examined whether G3BP1-positive granular structureswere formed during infections by a series of C knocked-out and mutated SeV recombinants and the parental Z strainby immunofluorescence microscopy (Figure 1; SupplementaryFigure S1).

Treating HeLa cells with sodium arsenite (NaAsO2) causedthe formation of granular structures in the cytosol, which were

Frontiers in Microbiology | www.frontiersin.org 3 August 2015 | Volume 6 | Article 804

Yoshida et al. Selective accumulation of IFN-β-inducing viral RNA

FIGURE 1 | Subcellular distribution of G3BP1 and TIAR in HeLa cellstreated with or without sodium arsenite (A), and infected with aseries of C-deficient rSeVs, dY1, dY2, d2Y, C′/C(-), and 4C(-), and aV-deficient rSeV, V(-) (B). HeLa cells treated with ethanol (EtOH) or sodiumarsenite (NaAsO2) or infected with the indicated virus were immunostainedwith anti-G3BP1 mAb, anti-TIAR mAb, and anti-SeV pAb. (C) The G3BP1foci in the samples of (B) were observed under a fluorescent microscope,

and the number of G3BP1 granule-positive cells was counted and ispresented as percentages against the number of SeV antigen-positive cells(closed bars). The relative amounts of IFN-β mRNA in cells infected with theindicated viruses were determined by qRT-PCR, and normalized to those ofβ-actin mRNA (open bars). (D) Similar experiments were performed for aseries of C-mutated SeV recombinants, and the results are presented as bargraphs, similarly to (C).

defined as SGs based on the expression of related proteins suchas G3BP1 and TIAR. G3BP1 and TIAR are well-establishedSG-associated proteins that are typically and diffusely presentthroughout the cytoplasm and dominantly present in the nucleus,respectively. However, treating cells with arsenite markedly

changed the localization to form SGs containing these proteinsin nearly all cells (Figure 1A).

When cells were infected with 4C(-), G3BP1-positive granularstructures were observed in almost 90% of SeV antigen-positivecells, and were considered to be SG-like structures since TIAR

Frontiers in Microbiology | www.frontiersin.org 4 August 2015 | Volume 6 | Article 804

Yoshida et al. Selective accumulation of IFN-β-inducing viral RNA

FIGURE 2 | Subcellular distribution of G3BP1 after the treatmentwith arsenite (A) and the infections with the virulent andavirulent NDV strains, MYD and D26, respectively (B), in HeLacells that received an empty pCAGGS.MCS vector or pCAGGS-C.

(C,D) Subcellular distribution of G3BP1 after the treatment witharsenite in HeLa cells infected with SeV-Z-WT (C) or 4C(-) (D). Cellswere immunostained with anti-G3BP1 mAb together with anti-SeV,anti-NDV, or anti-SeV C pAbs.

was also detected in the majority of the granules (Figures 1B,C).In this situation, TIAR was mostly in the cytoplasmic structures,whereas a larger part of TIAR was still observed in the nucleusin the arsenite-treated cells. This difference is probably dueto the different exposure time to the stimuli: 30 min. for thetreatment with arsenite and 24 h for the infection. In contrast,the percentages were only 8% or less in cells infected with the

parental Z-WT as well as the dY1, dY2, and d2Y recombinants,which lacked the smaller C proteins, Y1, Y2, and both Y1 andY2, respectively (Figure 1C; Supplementary Figure S1A). Aninfection by C′/C(-) and V(-), lacking the larger C proteins,C′ and C, and the V protein, respectively, resulted in a slightincrease in the number of granules (Figure 1C; SupplementaryFigure S1A). Of note, unlike the viruses reported previously,

Frontiers in Microbiology | www.frontiersin.org 5 August 2015 | Volume 6 | Article 804

Yoshida et al. Selective accumulation of IFN-β-inducing viral RNA

such as NS1-deficient IAV and vesicular stomatitis virus (Moket al., 2012; Onomoto et al., 2012; Dinh et al., 2013), thefluorescence of the SeV antigen was not colocalized with that ofthe representative SG marker G3BP1 in the granules (Figure 1B;Supplementary Figure S1).

IFN-β mRNA levels in the infected cells were also comparedbetween the viruses (Figure 1C). Strong correlations wereobserved between IFN-β mRNA levels and the percentages ofgranular structure-forming cells against infected cells. Similarstrong correlations were observed for a series of C mutant virusesthat possessed single- to triple-amino-acid substitutions of highlyconserved, charged amino acids within the C proteins, whichdiminished their ability to antagonize the host IFN system tovarious degrees (Figure 1D; Supplementary Figure S1B; Irie et al.,2010).

The marked difference observed in granular formation andIFN-β mRNA levels among the viruses was not due to theirdifferent growing abilities. The one-step growth kinetics of allthe recombinants used above was previously reported to besimilar to that of the parental Z-WT, except for 4C(-) andC′/C(-), the titers of which were 1–2 logs lower than that ofZ-WT throughout the time course, and viral protein synthesisin the infected cells did not significantly differ among theviruses examined (Kurotani et al., 1998; Irie et al., 2008a,2010). Treatment of HeLa cells with IFN-α did not induceG3BP1-positive granules, indicating that the formation of SG-like structures observed by SeV infection was triggered by SeVinfection, but not by SeV-induced type I IFNs (SupplementaryFigure S1C).

Taken together, these results strongly suggested that arelationship may exist between the formation of G3BP1-positivegranules and the induction of IFN-β in the C recombinants, andalso that C proteins may suppress the formation of granules.

SeV could not Inhibit the Arsenite- andVirus-Triggered Formation of G3BP1-PositiveGranulesWe examined whether expression of the C protein aloneand infection of the non-granule-forming SeV could inhibitthe formation of granules in cells treated with arsenite orinfected with NDV (Figure 2; Supplementary Figure S2).NDV, another prototypic paramyxovirus, which was consideredto be an IFN-β-inducing virus, could induce G3BP1-positivegranules in nearly all cells infected with virulent as wellas avirulent strains (Miyadera and D26 strains, respectively;Figure 2B).

The C protein alone failed to inhibit the formation ofgranules in cells treated with arsenite (Figure 2A) as wellas infected with NDV (Figure 2B), although the number ofgranules was slightly reduced in the C-expressing cells treatedwith arsenite compared to that observed in the neighboringnon-C-expressing cells (Figure 2A). In addition to C, the otherP gene products, P and V proteins, also failed to inhibit theformation of both types of granule (Supplementary Figure S2).Infection by SeV-Z-WT and C-deficient recombinant 4C(-) alsofailed to inhibit the arsenite-induced formation of granules(Figures 2C,D). Small granules dispersed in the cytoplasm were

FIGURE 3 | The percentages of G3BP-positive granules against SeVantigen-positive cells and the relative amounts of IFN-β mRNA arepresented as bar graphs, similarly to Figure 1C.

induced by arsenite in both cases of infection by WT and 4C(-).Of note, the G3BP1-positive granules induced by arsenite andviruses, such as 4C(-) and NDV, differed in size; the granulesinduced by the infection were apparently larger than thoseinduced by arsenite (Figures 1 and 2), implying a possibledifference in cellular pathways leading to these two types ofgranular structure. These results demonstrated that SeV didnot have the ability to inhibit the formation of both types ofgranules.

The Induction of IFN-β in Infections by SeVCantell Strain was not Related to theFormation of SG-Like StructuresAmarked difference was noted in the abilities of the SeV strains toinduce IFN-β. Although most of the SeV strains including Z havebeen characterized by their strong ability to counteract the innateimmune system, the CNT strain has been widely used as a virusthat induces high levels of IFN-β (Baum et al., 2010; Tapia et al.,2013). Therefore, we compared the abilities of some SeV strainsto induce IFN-β and SG-like structures (Figure 3; SupplementaryFigure S3).

In all cases, G3BP1-positive granular structures were onlydetected in less than 9% of SeV antigen-positive cells (Figure 3;Supplementary Figure S3A). IFN-β mRNA was not highlyinduced in cells infected with the SeV strains, except for CNT,whereas CNT induced IFN-β mRNA at a level that was 46-foldhigher than that by Z (Figure 3). Regarding the SeV strains,unlike that observed in the C recombinants, a correlation wasnot observed between IFN-β mRNA levels and the percentage of

Frontiers in Microbiology | www.frontiersin.org 6 August 2015 | Volume 6 | Article 804

Yoshida et al. Selective accumulation of IFN-β-inducing viral RNA

FIGURE 4 | Continued

Frontiers in Microbiology | www.frontiersin.org 7 August 2015 | Volume 6 | Article 804

Yoshida et al. Selective accumulation of IFN-β-inducing viral RNA

FIGURE 4 | Continued

(A) The percentages of G3BP1 granule-positive cells against the total numberof cells that received total RNA samples prepared from HeLa cells treated withor without arsenite or infected with SeV-Z-WT, SeV-CNT, SeV-4C(-), orNDV-MYD in an average of 20 random microscopic fields are presented asbar graphs (closed bars). The calculated P values were: arsenite, P = 0.49;Z-WT, P = 0.15; CNT, P = 0.031; 4C(-), P = 0.0023; NDV, P = 0.0055. Therelative amounts of IFN-β mRNA in HeLa cells that received the RNA sampleswere determined by qRT-PCR, and normalized by those of β-actin mRNA.(B) The amounts of cbDI genomes as well as genome-length viral RNAs inHeLa cells infected with the SeVs, Z-WT, 4C(-), CNT, or CNT-lowDI, weredetermined by qRT-PCR. The relative ratio of the cbDI genomes to thegenome-length viral RNAs are presented as bar graphs. The relative amountsof IFN-β mRNA in HeLa cells that infected with the indicated viruses weredetermined, as performed in (A). (C) RNA samples were prepared from theviral protein-removed cell lysate and introduced into HeLa cells. Percentagesof G3BP1 granule-positive cells against the total number of cells in an averageof twenty random microscopic fields are presented as bar graphs, asperformed in (A). The calculated P values were: arsenite, P = 0.97; Z-WT,P = 0.96; CNT, P = 0.42; 4C(-), P = 0.033; NDV, P = 0.027. ∗P > 0.1;∗∗P < 0.05; ∗∗∗P < 0.01. The relative amounts of IFN-β mRNA in HeLa cellsthat received the RNA samples were determined, as performed in (A).

granular structure-forming cells against infected cells (Figure 3).The different levels of IFN-β mRNA induced by the strains couldnot be attributed to differences in viral growth (SupplementaryFigure S3B).

These results suggested that the formation of G3BP1-positivegranules was not necessarily required to sense the SeV CNTinfection, followed by the production of IFN-β, unlike the Crecombinants.

Non-Encapsidated Viral RNA Species, but notthe Encapsidated cbDI Genome, was a PotentInducer of G3BP1-Positive GranulesWe attempted to identify the reason for the difference ingranular formation between these two types of IFN-β-inducingvirus, 4C(-) and CNT. To address this issue, we first examinedthe ability of total RNA prepared from virus-infected as wellas arsenite-treated cells to induce G3BP1-positive granules(Figure 4A; Supplementary Figure S4A). Total RNA samplesprepared from cells infected with any of the IFN-β-inducingviruses, SeV-CNT, 4C(-), and NDV, were able to induce thegranules as well as IFN-β inHeLa cells despite no formation of thegranules by the CNT infection, whereas those from cells infectedwith the IFN-β-non-inducing SeV-Z-WT or treated with arsenitewere not (Figure 4A).

We then examined the content rates of cbDI genomes, a potentligand for RIG-I, against viral genome-length RNAs in the RNAsamples used above (Figure 4B). The CNT sample containedcbDI genomes at a level that was ∼15,700-fold higher than thatin the Z-WT sample, although the sample of 4C(-) containedsimilar levels to the Z-WT sample (Figure 4B). We furtherexamined the effect of removal of encapsidated viral RNA species,such as viral full-length and DI genomes, by three consecutiverounds of immunoprecipitation using antisera against the wholevirions of SeV and NDV prior to the preparation of RNA samples(Supplementary Figure S4B). Total RNA prepared from the post-immunoprecipitation samples of 4C(-) and NDV was still able

FIGURE 5 | (A) dsRNA was not detected by the anti-dsRNA antibody, J2.HeLa cells uninfected, (-), or infected with Z-WT or CNT were immunostainedwith the J2 antibody. (B,C) Subcellular distribution of J2-detectable dsRNAand G3BP1 (B) or RIG-I and G3BP1 (C) in HeLa cells infected with 4C(-) orNDV-MYD. Cells were immunostained with the J2 antibody and anti-G3BP1pAb (B) or anti-RIG-I pAb and anti-G3BP1 mAb (C). Arrowheads in (B)indicate the sites of G3BP1-positive granules. The signals of the J2 and RIG-Iantibodies were amplified using a TSA kit.

to induce the granules as well as IFN-β, but that of CNT lostthese abilities (Figure 4C). Since the naked cbDI genomes havebeen reported readily to form an ideal structure as the RIG-I ligands of 5′-triphosphated, blunt-ended dsRNA (Kolakofsky,1976), these results indicated that the major IFN-β-inducingviral RNA species produced in the cells infected with CNT wasencapsidated cbDI genomes, whereas those for SeV-4C(-) andNDV were not.

SeV-Induced G3BP1-Positive GranulesIncluded RIG-I, but not J2-Detectable ViraldsRNAThe results above suggested that there may be at least two typesof IFN-β-inducing viral RNA species with a marked difference in

Frontiers in Microbiology | www.frontiersin.org 8 August 2015 | Volume 6 | Article 804

Yoshida et al. Selective accumulation of IFN-β-inducing viral RNA

FIGURE 6 | Continued

FIGURE 6 | Continued

(A) FISH analysis of HeLa cells infected with the SeVs, Z-WT, 4C(-), CNT, andCNT-lowDI. HeLa cells were infected with the indicated viruses. After thefixation and permeabilization of cells, they underwent hybridization with anAlexa Fluor 488-labeled RNA probe complementary to the region of14,761–15,384 nt in the (-)-sense SeV genome, with or without being treatedwith Proteinase K at a final concentration of 50 μg/ml for 5 min. (B–D)Subcellular colocalization of FISH signals with SeV N (B), RIG-I (C), andG3BP1 (D). The FISH samples prepared in (A) were immunostained furtherwith an anti-SeV N mAb, anti-RIG-I pAb, or anti-G3BP1 mAb. Signals forRIG-I were amplified using the TSA kit. The arrowheads in (B) indicate thesites of FISH-positive inclusions.

the formation of SG-like granules. To elucidate this difference,we examined the subcellular distribution of unusual viral RNAspecies produced by 4C(-) and CNT (Figures 5 and 6).

We and other groups previously reported a strong correlationbetween the production of dsRNA detected by the anti-dsRNA antibody J2 (J2-dsRNA) and the induction of IFN-βin infections of the C-mutated SeV recombinants and NDV(Takeuchi et al., 2008; Irie et al., 2010). Therefore, we firstexamined the production and subcellular distribution of J2-dsRNA in the cells infected with the IFN-β-inducing viruses byimmunofluorescence microscopy (Figure 5). DsRNA fluorescentsignals were absent in cells infected with IFN-non-inducingZ-WT as well as in those with the IFN-β-inducing strain CNT(Figure 5A). In contrast, dsRNA fluorescent signals were clearlyobserved in the cytoplasm of cells infected with IFN-β-inducingSeV-4C(-) and NDV with dispersed and granular distributions,respectively (Figure 5B).

A previous study reported that NS1-deficient IAV infectionscaused RIG-I to form granular aggregates that contained SGmarkers as well as viral RNA (Onomoto et al., 2012). Therefore,the cells infected with 4C(-) and NDV were co-stained withanti-RIG-I and anti-G3BP1 antibodies (Figure 5C). Similar toIAV, RIG-I almost perfectly colocalized with the virus-inducedG3BP1-positive granules in both cases of infection (Figure 5C);however, unlike IAV, these granules did not colocalize with theviral antigen or viral J2-dsRNA (Figure 5B, arrowheads), asshown in Figures 1 and 2. Together with the results of Figure 4,J2-dsRNA appeared to be not or less encapsidated by viralnucleoproteins because the access of the antibody to and theformation of dsRNA by tightly encapsidated viral RNA, such asviral genomes, was unlikely.

Non- and Partially Encapsidated Unusual ViralRNA Species were Selectively Incorporatedinto avSG-Like and Non-avSG-Like InclusionsWe further attempted to visualize the subcellular distribution ofthe unusual viral RNA species that were not fully encapsidated,unlike the genome-length viral RNAs, by FISH analysis(Figure 6). Infected cell samples were prepared withouta protease treatment to exclude fully encapsidated viralRNA species, and were then stained with an RNA probecomplementary to the 14,761–15,384 region of the (-)-senseviral genome RNA, designed by referring to two well-characterized cbDI genomes of SeV (Calain et al., 1992;

Frontiers in Microbiology | www.frontiersin.org 9 August 2015 | Volume 6 | Article 804

Yoshida et al. Selective accumulation of IFN-β-inducing viral RNA

TABLE 1 | Properties of unusual viral RNA species produced by infections of the IFN-β-inducing SeVs.

Type SeV Defined as Detected by Colocalized with Encapsidatedwith N

Accumulated intocytoplasmic structures

Viral protein responsiblefor the production

J2 FISH SG markers SeV N RIG-I

I 4C(-) dsRNA + − − − − No No,dispersed in the cytoplasm

C

II 4C(-) Unknown − + + − + No Yes,avSG-like structures

C

III CNT cbDI − + − + − Yes,but not fully

Yes,undefined, non-avSGstructures

ND

ND, not determined.

Martinez-Gil et al., 2013). Fluorescence-positive cytoplasmicinclusions were observed in the 4C(-)- as well as in the CNT-infected samples, while an apparent signal was not detectedin Z-WT-infected or uninfected samples (Figure 6A). Whenthe Z-WT-infected sample was treated with Proteinase K,apparent signals were observed throughout the cytoplasm(Figure 6A). These results strongly suggested that fullyencapsidated viral genomes were not detected, whereas non-or partially encapsidated viral RNA species were detectable inthis system. The FISH-positive inclusions observed in CNT-infected cells were no longer detected in the cells infectedwith the CNT-lowDI (Figure 6A), which contained ∼10-foldfewer cbDI genomes than the original CNT sample (Figure 4B).Together with the results of Figure 4, the FISH signalsobserved in the CNT-infected cells were considered as the cbDIgenomes.

The FISH samples of 4C(-) and CNT-infected cells werefurther immunostained to examine the subcellular colocalizationof FISH signals, the SeV N protein, RIG-I, and G3BP1(Figures 6B–D, respectively). The SeV N protein wasdetected in the FISH-positive inclusions of CNT-infectedcells, suggesting that the RNA species detected in the CNTsamples were only partially, not fully, encapsidated, unlikethe fully encapsidated, full-length, intact viral genomes(Figure 6B). In contrast, the N protein was not detected inthe inclusions of 4C(-)-infected cells, which suggested that theRNA species detected in the 4C(-) samples were not encapsidated(Figure 6B). The FISH-positive inclusions of CNT-infectedcells were not apparently colocalized with RIG-I or G3BP1,whereas most of those in the 4C(-)-infected cells colocalizedobviously with RIG-I and G3BP1, unlike the J2-dsRNA(Figures 6C,D).

These results indicated that unusual viral RNA speciesharboring the 5′-region of (-)-sense SeV genome RNA, whichwas not produced during infection by IFN-β-non-inducingZ-WT, was produced in infections by IFN-β-inducing CNTand 4C(-), and were selectively formed into distinct cytoplasmicinclusions in an RNA-type-dependent manner. The inclusionsin the 4C(-)-infected cells could be identified as avSGs, andmay be the site to detect viral RNA by RIG-I, whereas thosein CNT-infected cells were not avSGs, but as-yet-unidentifiedstructures.

Discussion

Although a number of studies have examined host innateimmune responses against pathogenic microbes including RNAviruses, the virus-derived RNA species that serve as PAMPs inreal viral infections and the sites at which PAMPs are recognizedby RLRs have yet to be clarified in detail. Two important insightswere reported recently. Regarding the real PAMPs in infections byRNA viruses, the cb and id types of DI genomes were identified asthe ligands of RIG-I in SeV- and IAV-infected cells, respectively,both of which could form ideal structures as the RIG-I ligands(Baum et al., 2010; Baum and Garcia-Sastre, 2011; Martinez-Gil et al., 2013). The SG-like structures have been suggested toserve as the sites at which the RLRs encounter viral RNA andsubsequently activate the IFN signaling pathways in infections byRNA viruses (Onomoto et al., 2012; Yoo et al., 2014). In orderto establish what and where viral RNA species were detectedby RLRs, in the present study, we compared two types of IFN-β-inducing SeV, a recombinant 4C(-) and a strain CNT, with anon-IFN-β-inducing Z strain, in terms of the formation of SG-like granules and the production of unusual viral RNA species.A major advantage of our study is that the comparison can beperformed within the context of the same viral species.

Several types of unusual viral RNA species were found tobe generated in cells infected with the IFN-inducing SeVs butnot those with the IFN-non-inducing SeVs (summarized inTable 1). One was a dsRNA (J2-dsRNA) that was detected bythe anti-dsRNA antibody, J2 (type I in Table 1). As for SeV,the generation of J2-dsRNA may have been restricted by theC proteins because it was only detected in cells infected withC-deficient or mutated recombinants, but not in those with intactSeVs (Figure 5; Takeuchi et al., 2008; Irie et al., 2010). Weand other groups demonstrated that J2-dsRNA activated PKR,and this was followed by the phosphorylation of eIF2 and theproduction of IFN-β, both of which resulted in antiviral effectsin the host cells. The activation of PKR has been reported toinduce the formation of SG-like structures during infections bysome RNA viruses, such as IAV and measles virus (MeV; Moket al., 2012; Onomoto et al., 2012; Okonski and Samuel, 2013),and this also appears to be the case for the SeV C recombinants.Unlike these viruses, the SG-like structures formed during 4C(-)infections did not include the J2-dsRNA, which was dispersed in

Frontiers in Microbiology | www.frontiersin.org 10 August 2015 | Volume 6 | Article 804

Yoshida et al. Selective accumulation of IFN-β-inducing viral RNA

the cytoplasm, although they contained RIG-I (Figure 5). Thismay lead to the assertion that the SG-like structures inducedduring infections by C recombinants are not the sites at whichto detect SeV infections.

However, the SG-like structures formed by 4C(-) wererevealed to contain another type of unusual viral RNA speciesby FISH analysis, in which an RNA probe targeting the 600 ntregion of the 5′-end of the (-)-sense SeV genome was used(type II in Table 1; Figure 6). This now strongly suggeststhat the SG-like structures found in the 4C(-) infection aredefined as avSGs. Unlike the IAV infection, the type I andII RNA species seemed to be not or less encapsidated, giventhat they were not colocalized with viral antigens and were notremoved from the infected cell lysates by immunoprecipitationusing anti-SeV antibody (Figures 1 and 4–6). SeV trailer RNA,which is transcribed from the 3′-ends of (+)-sense antigenomeRNA, was previously reported to interact with TIAR to inhibitapoptosis and the formation of SGs induced by infection (Iseniet al., 2002). The 4C(-) virus was shown to induce apoptosismore quickly and severely in infected cells than the WT virus(Irie et al., 2010). Although appearing to contain the non-encapsidated 5′-end of a (-)-sense genome RNA, at least inthe part that is concordant with the trailer RNA, it is unlikelythat the type II RNA observed in the 4C(-)-infected cells hasthe ability of the trailer RNA to inhibit apoptosis and SGformation. Although details of the type I and II RNAs remainto be solved, given the cytoplasmic replication of SeV RNAwithout forming inclusion bodies and the differences of the RNAspecies in subcellular distribution and reactivity with the J2 andFISH probe, the type I dsRNA somewhat unwound actively orincidentally into the type II RNA might be accumulated into theavSGs.

Although the SG-like structures and J2-dsRNA were notdetected during CNT infections, FISH-positive non-granular-shaped inclusions were observed (type III in Table 1; Figure 6).These CNT inclusions were markedly different from thoseof 4C(-). The inclusions did not include RIG-I or G3BP1, butcontained the N protein (Figure 6), suggesting that the inclusionswere not avSGs, and that the type III RNA was at least partiallyencapsidated. The CNT stock used in the present study containeda larger amount of cbDI genomes than the other viral stocks(Figure 4B). The SeV cbDI genomes have been shown to bepartially and/or more loosely encapsidated than intact genomes(Kolakofsky, 1976; Strahle et al., 2006). The SeV cbDI genomeswere recently identified as strong ligands for RIG-I (Baum et al.,2010; Tapia et al., 2013). Indeed, the CNT-lowDI had lost theability to induce IFN-β and the inclusions had not been found inthe infected cells (Figures 4B and 6). Taken together, the type IIIRNA was identified as the cbDI genome. These results indicatedthat the process of detecting infections and the subsequentinduction of IFN-β differed largely between 4C(-) and CNT:avSG-dependent and -independent mechanisms for 4C(-) andCNT, respectively.

Most viruses have been shown to possess the ability toantagonize host IFN pathways in order to avoid activatinghost antiviral actions, and this strategy is mostly based on acounteraction against the molecules involved in these pathways

(Versteeg and Garcia-Sastre, 2010). However, a recent studyreported that encephalomyocarditis virus (EMCV) has the abilityto disrupt SGs by cleaving G3BP1 in order to avoid the innateimmune detection of its infection and subsequent inductionof IFN-β (Ng et al., 2013), suggesting that another effectivestrategy for viral evasion from the IFN system by preventingavSG formation exists. Generation of the unusual viral RNAspecies triggering the production of IFN-β seems to be suppressedduring intact RNA viral replication. Similarly to SeV, anotherparamyxovirus MeV C protein was recently shown to impairthe production of J2-dsRNA and the activation of PKR coupledwith the formation of SG-like structures (Okonski and Samuel,2013; Pfaller et al., 2014). Unlike the case of SeV, in which theknockout of C resulted in the production of J2-dsRNA (Figures 5and 6), but not cbDI genomes, the J2-dsRNA produced duringthe infection by a C-deficient MeV recombinant was reportedto be a cbDI genome (Pfaller et al., 2014). Although the cbDIgenomes were more dominantly produced by SeV-CNT thanby the other strains and C-recombinants tested (Figure 4Band data not shown), this unique property of CNT mightbe attributed to its C protein that possibly have a functionaldifference with those of the other SeVs. The C proteins ofboth SeV and MeV have been shown to play critical roles inmodulating viral RNA synthesis and maintaining its integrity bypotentially stabilizing the ribonucleoprotein (RNP)-polymerasecomplex (Tapparel et al., 1997; Reutter et al., 2001; Bankamp et al.,2005; Irie et al., 2008a, 2014; Ito et al., 2013). Dysfunctions in theC proteins may result in a disturbance in integrity, leading to theproduction of the unusual, IFN-β-inducing RNA species.

Conclusion

The results of the present study indicate that several types ofIFN-β-inducible, unusual viral RNA species may be producedduring SeV infections and included in avSG-like and non-avSG-like cytoplasmic inclusions, which suggests that RNA-type-dependent mechanisms recognize and accumulate such unusualviral RNAs in specific compartments. In addition, the productionof these unusual RNA species may be restricted during intact viralreplication in order to avoid detection by host innate immunity.

Acknowledgments

We thank the staff of the Analysis Center of Life Science,Hiroshima University, for the use of their facilities. We also thankDr. K. Takeuchi (University of Tsukuba) for fruitful discussions.This work was supported by JSPS KAKENHI (Grant Numbers23790505 and 25460569).

Supplementary Material

The Supplementary Material for this article can be foundonline at: http://journal.frontiersin.org/article/10.3389/fmicb.2015.00804

Frontiers in Microbiology | www.frontiersin.org 11 August 2015 | Volume 6 | Article 804

Yoshida et al. Selective accumulation of IFN-β-inducing viral RNA

References

Akira, S., Uematsu, S., and Takeuchi, O. (2006). Pathogen recognition and innateimmunity. Cell 124, 783–801. doi: 10.1016/j.cell.2006.02.015

Anderson, P., and Kedersha, N. (2002). Visibly stressed: the role of eIF2, TIA-1,and stress granules in protein translation. Cell Stress Chaperones 7, 213–221.doi: 10.1379/1466-1268(2002)007<0213:VSTROE>2.0.CO;2

Anderson, P., and Kedersha, N. (2006). RNA granules. J. Cell Biol. 172, 803–808.doi: 10.1083/jcb.200512082

Ariumi, Y., Kuroki, M., Kushima, Y., Osugi, K., Hijikata, M.,Maki, M., et al. (2011).Hepatitis C virus hijacks P-body and stress granule components around lipiddroplets. J. Virol. 85, 6882–6892. doi: 10.1128/JVI.02418-10

Balagopal, V., and Parker, R. (2009). Polysomes, P bodies and stress granules:states and fates of eukaryotic mRNAs. Curr. Opin. Cell Biol. 21, 403–408. doi:10.1016/j.ceb.2009.03.005

Bankamp, B., Wilson, J., Bellini, W. J., and Rota, P. A. (2005). Identification ofnaturally occurring amino acid variations that affect the ability of the measlesvirus C protein to regulate genome replication and transcription. Virology 336,120–129. doi: 10.1016/j.virol.2005.03.009

Baum, A., and Garcia-Sastre, A. (2011). Differential recognition of viral RNA byRIG-I. Virulence 2, 166–169. doi: 10.4161/viru.2.2.15481

Baum, A., Sachidanandam, R., and Garcia-Sastre, A. (2010). Preference ofRIG-I for short viral RNA molecules in infected cells revealed by next-generation sequencing. Proc. Natl. Acad. Sci. U.S.A. 107, 16303–16308. doi:10.1073/pnas.1005077107

Bowzard, J. B., Ranjan, P., and Sambhara, S. (2013). RIG-I goes beyondnaked recognition. Cell Host Microbe 13, 247–249. doi: 10.1016/j.chom.2013.02.012

Buchan, J. R., and Parker, R. (2009). Eukaryotic stress granules: the ins and outs oftranslation. Mol. Cell 36, 932–941. doi: 10.1016/j.molcel.2009.11.020

Calain, P., Curran, J., Kolakofsky, D., and Roux, L. (1992). Molecular cloningof natural paramyxovirus copy-back defective interfering RNAs and theirexpression from DNA. Virology 191, 62–71. doi: 10.1016/0042-6822(92)90166-M

Carroll, K., Hastings, C., and Miller, C. L. (2014). Amino acids 78 and 79 ofMammalian Orthoreovirus protein microNS are necessary for stress granulelocalization, core protein lambda2 interaction, and de novo virus replication.Virology 448, 133–145. doi: 10.1016/j.virol.2013.10.009

Davis, W. G., Bowzard, J. B., Sharma, S. D., Wiens, M. E., Ranjan, P.,Gangappa, S., et al. (2012). The 3’ untranslated regions of influenza genomicsequences are 5’PPP-independent ligands for RIG-I. PLoS ONE 7:e32661. doi:10.1371/journal.pone.0032661

Dinh, P. X., Beura, L. K., Das, P. B., Panda, D., Das, A., and Pattnaik, A. K. (2013).Induction of stress granule-like structures in vesicular stomatitis virus-infectedcells. J. Virol. 87, 372–383. doi: 10.1128/JVI.02305-12

Fitzgerald, K. D., and Semler, B. L. (2013). Poliovirus infection induces the co-localization of cellular protein SRp20 with TIA-1, a cytoplasmic stress granuleprotein. Virus Res. 176, 223–231. doi: 10.1016/j.virusres.2013.06.012

Fujii, Y., Sakaguchi, T., Kiyotani, K., Huang, C., Fukuhara, N., Egi, Y., et al. (2002).Involvement of the leader sequence in Sendai virus pathogenesis revealed byrecovery of a pathogenic field isolate from cDNA. J. Virol. 76, 8540–8547. doi:10.1128/JVI.76.17.8540-8547.2002

Garaigorta, U., Heim, M. H., Boyd, B., Wieland, S., and Chisari, F. V. (2012).Hepatitis C virus (HCV) induces formation of stress granules whose proteinsregulate HCV RNA replication and virus assembly and egress. J. Virol. 86,11043–11056. doi: 10.1128/JVI.07101-11

Gitlin, L., Barchet, W., Gilfillan, S., Cella, M., Beutler, B., Flavell, R. A.,et al. (2006). Essential role of mda-5 in type I IFN responses topolyriboinosinic:polyribocytidylic acid and encephalomyocarditis picornavirus.Proc. Natl. Acad. Sci. U.S.A. 103, 8459–8464. doi: 10.1073/pnas.0603082103

Hausmann, S., Marq, J. B., Tapparel, C., Kolakofsky, D., and Garcin, D. (2008).RIG-I and dsRNA-induced IFNbeta activation. PLoS ONE 3:e3965. doi:10.1371/journal.pone.0003965

Holcik, M., and Sonenberg, N. (2005). Translational control in stress and apoptosis.Nat. Rev. Mol. Cell Biol. 6, 318–327. doi: 10.1038/nrm1618

Hornung, V., Ellegast, J., Kim, S., Brzozka, K., Jung, A., Kato, H., et al. (2006).5’-Triphosphate RNA is the ligand for RIG-I. Science 314, 994–997. doi:10.1126/science.1132505

Irie, T., Kiyotani, K., Igarashi, T., Yoshida, A., and Sakaguchi, T. (2012). Inhibitionof interferon regulatory factor 3 activation by paramyxovirus V protein. J. Virol.86, 7136–7145. doi: 10.1128/JVI.06705-11

Irie, T., Nagata, N., Igarashi, T., Okamoto, I., and Sakaguchi, T. (2010).Conserved charged amino acids within Sendai virus C protein play multipleroles in the evasion of innate immune responses. PLoS ONE 5:e10719. doi:10.1371/journal.pone.0010719

Irie, T., Nagata, N., Yoshida, T., and Sakaguchi, T. (2008a). Paramyxovirus Sendaivirus C proteins are essential for maintenance of negative-sense RNA genomein virus particles. Virology 374, 495–505. doi: 10.1016/j.virol.2008.01.004

Irie, T., Nagata, N., Yoshida, T., and Sakaguchi, T. (2008b). Recruitmentof Alix/AIP1 to the plasma membrane by Sendai virus C proteinfacilitates budding of virus-like particles. Virology 371, 108–120. doi:10.1016/j.virol.2007.09.020

Irie, T., Okamoto, I., Yoshida, A., Nagai, Y., and Sakaguchi, T. (2014).Sendai virus C proteins regulate viral genome and antigenome synthesis todictate the negative genome polarity. J. Virol. 88, 690–698. doi: 10.1128/JVI.02798-13

Irie, T., Yoshida, A., and Sakaguchi, T. (2013). Clustered basic amino acids of thesmall sendai virus C proteinY1 are critical to its RANGTPase-mediated nuclearlocalization. PLoS ONE 8:e73740. doi: 10.1371/journal.pone.0073740

Iseni, F., Garcin, D., Nishio, M., Kedersha, N., Anderson, P., and Kolakofsky, D.(2002). Sendai virus trailer RNA binds TIAR, a cellular protein involvedin virus-induced apoptosis. EMBO J. 21, 5141–5150. doi: 10.1093/emboj/cdf513

Ito, M., Iwasaki, M., Takeda, M., Nakamura, T., Yanagi, Y., and Ohno, S.(2013). Measles virus nonstructural C proteinmodulates viral RNA polymeraseactivity by interacting with host protein SHCBP1. J. Virol. 87, 9633–9642. doi:10.1128/JVI.00714-13

Kaisho, T., and Akira, S. (2006). Toll-like receptor function and signaling.J. Allergy Clin. Immunol. 117, 979–987; quiz 988. doi: 10.1016/j.jaci.2006.02.023

Kato, A., Kiyotani, K., Sakai, Y., Yoshida, T., and Nagai, Y. (1997). Theparamyxovirus, Sendai virus, V protein encodes a luxury function required forviral pathogenesis. EMBO J. 16, 578–587. doi: 10.1093/emboj/16.3.578

Kato, A., Sakai, Y., Shioda, T., Kondo, T., Nakanishi, M., and Nagai, Y. (1996).Initiation of Sendai virus multiplication from transfected cDNA or RNAwith negative or positive sense. Genes Cells 1, 569–579. doi: 10.1046/j.1365-2443.1996.d01-261.x

Kato, H., Takahasi, K., and Fujita, T. (2011). RIG-I-like receptors: cytoplasmicsensors for non-self RNA. Immunol. Rev. 243, 91–98. doi: 10.1111/j.1600-065X.2011.01052.x

Kedersha, N., and Anderson, P. (2002). Stress granules: sites of mRNA triage thatregulate mRNA stability and translatability. Biochem. Soc. Trans. 30, 963–969.

Khaperskyy, D. A., Hatchette, T. F., and Mccormick, C. (2012). Influenza A virusinhibits cytoplasmic stress granule formation. FASEB J. 26, 1629–1639. doi:10.1096/fj.11-196915

Kiyotani, K., Takao, S., Sakaguchi, T., and Yoshida, T. (1990). Immediate protectionof mice from lethal wild-type Sendai virus (HVJ) infections by a temperature-sensitive mutant, HVJpi, possessing homologous interfering capacity. Virology177, 65–74. doi: 10.1016/0042-6822(90)90460-9

Kolakofsky, D. (1976). Isolation and characterization of Sendai virus DI-RNAs.Cell8, 547–555. doi: 10.1016/0092-8674(76)90223-3

Kurotani, A., Kiyotani, K., Kato, A., Shioda, T., Sakai, Y., Mizumoto, K., et al.(1998). Sendai virus C proteins are categorically nonessential gene products butsilencing their expression severely impairs viral replication and pathogenesis.Genes Cells 3, 111–124. doi: 10.1046/j.1365-2443.1998.00170.x

Lindquist, M. E., Lifland, A. W., Utley, T. J., Santangelo, P. J., and Crowe, J. E. Jr.(2010). Respiratory syncytial virus induces host RNA stress granules to facilitateviral replication. J. Virol. 84, 12274–12284. doi: 10.1128/JVI.00260-10

Lindquist, M. E., Mainou, B. A., Dermody, T. S., and Crowe, J. E. Jr. (2011).Activation of protein kinase R is required for induction of stress granules byrespiratory syncytial virus but dispensable for viral replication. Virology 413,103–110. doi: 10.1016/j.virol.2011.02.009

Marq, J. B., Hausmann, S., Veillard, N., Kolakofsky, D., and Garcin, D. (2011).Short double-stranded RNAs with an overhanging 5’ ppp-nucleotide, as foundin arenavirus genomes, act as RIG-I decoys. J. Biol. Chem. 286, 6108–6116. doi:10.1074/jbc.M110.186262

Frontiers in Microbiology | www.frontiersin.org 12 August 2015 | Volume 6 | Article 804

Yoshida et al. Selective accumulation of IFN-β-inducing viral RNA

Martinez-Gil, L., Goff, P. H., Hai, R., Garcia-Sastre, A., Shaw, M. L., and Palese, P.(2013). A Sendai virus-derived RNA agonist of RIG-I as a virus vaccineadjuvant. J. Virol. 87, 1290–1300. doi: 10.1128/JVI.02338-12

McInerney, G. M., Kedersha, N. L., Kaufman, R. J., Anderson, P., and Liljestrom, P.(2005). Importance of eIF2alpha phosphorylation and stress granule assemblyin alphavirus translation regulation. Mol. Biol. Cell 16, 3753–3763. doi:10.1091/mbc.E05-02-0124

Mok, B.W., Song,W.,Wang, P., Tai, H., Chen, Y., Zheng, M., et al. (2012). The NS1protein of influenza A virus interacts with cellular processing bodies and stressgranules through RNA-associated protein 55 (RAP55) during virus infection.J. Virol. 86, 12695–12707. doi: 10.1128/JVI.00647-12

Ng, C. S., Jogi, M., Yoo, J. S., Onomoto, K., Koike, S., Iwasaki, T., et al. (2013).Encephalomyocarditis virus disrupts stress granules, the critical platform fortriggering antiviral innate immune responses. J. Virol. 87, 9511–9522. doi:10.1128/JVI.03248-12

Okonski, K. M., and Samuel, C. E. (2013). Stress granule formation induced bymeasles virus is protein kinase PKR dependent and impaired by RNA adenosinedeaminase ADAR1. J. Virol. 87, 756–766. doi: 10.1128/JVI.02270-12

Onomoto, K., Jogi, M., Yoo, J. S., Narita, R., Morimoto, S., Takemura, A.,et al. (2012). Critical role of an antiviral stress granule containing RIG-Iand PKR in viral detection and innate immunity. PLoS ONE 7:e43031. doi:10.1371/journal.pone.0043031

Pager, C. T., Schutz, S., Abraham, T. M., Luo, G., and Sarnow, P. (2013).Modulation of hepatitis C virus RNA abundance and virus release by dispersionof processing bodies and enrichment of stress granules. Virology 435, 472–484.doi: 10.1016/j.virol.2012.10.027

Panas, M. D., Varjak, M., Lulla, A., Eng, K. E., Merits, A., Karlsson Hedestam, G. B.,et al. (2012). Sequestration of G3BP coupled with efficient translation inhibitsstress granules in Semliki Forest virus infection. Mol. Biol. Cell 23, 4701–4712.doi: 10.1091/mbc.E12-08-0619

Pfaller, C. K., Radeke, M. J., Cattaneo, R., and Samuel, C. E. (2014). Measles virusC protein impairs production of defective copyback double-stranded viral RNAand activation of protein kinase R. J. Virol. 88, 456–468. doi: 10.1128/JVI.02572-13

Pichlmair, A., Schulz, O., Tan, C. P., Naslund, T. I., Liljestrom, P., Weber, F., et al.(2006). RIG-I-mediated antiviral responses to single-stranded RNA bearing5’-phosphates. Science 314, 997–1001. doi: 10.1126/science.1132998

Qin, Q., Carroll, K., Hastings, C., and Miller, C. L. (2011). Mammalianorthoreovirus escape from host translational shutoff correlates with stressgranule disruption and is independent of eIF2alpha phosphorylation and PKR.J. Virol. 85, 8798–8810. doi: 10.1128/JVI.01831-10

Qin, Q., Hastings, C., and Miller, C. L. (2009). Mammalian orthoreovirus particlesinduce and are recruited into stress granules at early times postinfection. J. Virol.83, 11090–11101. doi: 10.1128/JVI.01239-09

Raaben, M., Groot Koerkamp, M. J., Rottier, P. J., and De Haan, C. A.(2007). Mouse hepatitis coronavirus replication induces host translationalshutoff and mRNA decay, with concomitant formation of stress granules andprocessing bodies.Cell. Microbiol. 9, 2218–2229. doi: 10.1111/j.1462-5822.2007.00951.x

Reutter, G. L., Cortese-Grogan, C., Wilson, J., and Moyer, S. A. (2001). Mutationsin the measles virus C protein that up regulate viral RNA synthesis. Virology285, 100–109. doi: 10.1006/viro.2001.0962

Ruggieri, A., Dazert, E., Metz, P., Hofmann, S., Bergeest, J. P., Mazur, J., et al.(2012). Dynamic oscillation of translation and stress granule formation markthe cellular response to virus infection. Cell Host Microbe 12, 71–85. doi:10.1016/j.chom.2012.05.013

Runge, S., Sparrer, K. M., Lassig, C., Hembach, K., Baum, A., Garcia-Sastre, A., et al.(2014). In vivo ligands of MDA5 and RIG-I inmeasles virus-infected cells. PLoSPathog. 10:e1004081. doi: 10.1371/journal.ppat.1004081

Saito, T., Hirai, R., Loo, Y. M., Owen, D., Johnson, C. L., Sinha, S. C., et al.(2007). Regulation of innate antiviral defenses through a shared repressordomain in RIG-I and LGP2. Proc. Natl. Acad. Sci. U.S.A. 104, 582–587. doi:10.1073/pnas.0606699104

Sakaguchi, T., Irie, T., Kuwayama, M., Ueno, T., Yoshida, A., and Kawabata, R.(2011). Analysis of interaction of Sendai virus V protein and melanoma

differentiation-associated gene 5. Microbiol. Immunol. 55, 760–767. doi:10.1111/j.1348-0421.2011.00379.x

Sakaguchi, T., Kato, A., Sugahara, F., Shimazu, Y., Inoue, M., Kiyotani, K., et al.(2005). AIP1/Alix is a binding partner of Sendai virus C protein and facilitatesvirus budding. J. Virol. 79, 8933–8941. doi: 10.1128/JVI.79.14.8933-8941.2005

Schmolke, M., Patel, J. R., De Castro, E., Sanchez-Aparicio, M. T., Uccellini, M. B.,Miller, J. C., et al. (2014). RIG-I detects mRNA of intracellular Salmonellaenterica serovar Typhimurium during bacterial infection. MBio 5, e01006–e01014. doi: 10.1128/mBio.01006-14

Strahle, L., Garcin, D., and Kolakofsky, D. (2006). Sendai virus defective-interferinggenomes and the activation of interferon-beta. Virology 351, 101–111. doi:10.1016/j.virol.2006.03.022

Strahle, L., Marq, J. B., Brini, A., Hausmann, S., Kolakofsky, D., and Garcin, D.(2007). Activation of the beta interferon promoter by unnatural Sendai virusinfection requires RIG-I and is inhibited by viral C proteins. J. Virol. 81,12227–12237. doi: 10.1128/JVI.01300-07

Takeuchi, K., Komatsu, T., Kitagawa, Y., Sada, K., and Gotoh, B. (2008). Sendaivirus C protein plays a role in restricting PKR activation by limiting thegeneration of intracellular double-stranded RNA. J. Virol. 82, 10102–10110. doi:10.1128/JVI.00599-08

Tapia, K., Kim, W. K., Sun, Y., Mercado-Lopez, X., Dunay, E., Wise, M., et al.(2013). Defective viral genomes arising in vivo provide critical danger signalsfor the triggering of lung antiviral immunity. PLoS Pathog. 9:e1003703. doi:10.1371/journal.ppat.1003703

Tapparel, C., Hausmann, S., Pelet, T., Curran, J., Kolakofsky, D., and Roux, L.(1997). Inhibition of Sendai virus genome replication due to promoter-increased selectivity: a possible role for the accessory C proteins. J. Virol. 71,9588–9599.

Toyoda, T., Sakaguchi, T., Imai, K., Inocencio, N. M., Gotoh, B., Hamaguchi, M.,et al. (1987). Structural comparison of the cleavage-activation site of the fusionglycoprotein between virulent and avirulent strains of Newcastle disease virus.Virology 158, 242–247. doi: 10.1016/0042-6822(87)90261-3

Versteeg, G. A., and Garcia-Sastre, A. (2010). Viral tricks to grid-lockthe type I interferon system. Curr. Opin. Microbiol. 13, 508–516. doi:10.1016/j.mib.2010.05.009

Weber,M., Gawanbacht, A., Habjan, M., Rang, A., Borner, C., Schmidt, A. M., et al.(2013). Incoming RNA virus nucleocapsids containing a 5’-triphosphorylatedgenome activate RIG-I and antiviral signaling. Cell Host Microbe 13, 336–346.doi: 10.1016/j.chom.2013.01.012

White, J. P., Cardenas, A. M., Marissen, W. E., and Lloyd, R. E. (2007). Inhibitionof cytoplasmic mRNA stress granule formation by a viral proteinase. Cell HostMicrobe 2, 295–305. doi: 10.1016/j.chom.2007.08.006

White, J. P., and Lloyd, R. E. (2011). Poliovirus unlinks TIA1 aggregationand mRNA stress granule formation. J. Virol. 85, 12442–12454. doi:10.1128/JVI.05888-11

Yoneyama, M., Kikuchi, M., Natsukawa, T., Shinobu, N., Imaizumi, T.,Miyagishi, M., et al. (2004). The RNA helicase RIG-I has an essential functionin double-stranded RNA-induced innate antiviral responses. Nat. Immunol. 5,730–737. doi: 10.1038/ni1087

Yoo, J. S., Takahasi, K., Ng, C. S., Ouda, R., Onomoto, K., Yoneyama, M.,et al. (2014). DHX36 enhances RIG-I signaling by facilitating PKR-mediated antiviral stress granule formation. PLoS Pathog. 10:e1004012. doi:10.1371/journal.ppat.1004012

Conflict of Interest Statement: The authors declare that the research wasconducted in the absence of any commercial or financial relationships that couldbe construed as a potential conflict of interest.

Copyright © 2015 Yoshida, Kawabata, Honda, Tomonaga, Sakaguchi and Irie. Thisis an open-access article distributed under the terms of the Creative CommonsAttribution License (CC BY). The use, distribution or reproduction in other forumsis permitted, provided the original author(s) or licensor are credited and that theoriginal publication in this journal is cited, in accordance with accepted academicpractice. No use, distribution or reproduction is permitted which does not complywith these terms.

Frontiers in Microbiology | www.frontiersin.org 13 August 2015 | Volume 6 | Article 804